Haier Leverages Model-Based Design to Accelerate Software Development for Smart Appliances - MATLAB & Simulink

Haier Leverages Model-Based Design to Accelerate Software Development for Smart Appliances

“With Model-Based Design, we can check the design results directly and quickly via simulation without any real hardware. Writing C code ourselves may require many breakpoints and intermediate variables. Automatic code generation not only minimizes human errors, but also saves more time for code review and inspection, which further improves efficiency.”

Key Outcomes

  • Simulating the system makes it easy to observe and understand whether the design is meeting requirements
  • Modeling and simulating the appliance system makes it easy to verify control algorithms before hardware becomes available
  • Generating code from the algorithm models reduces the cost of manual coding, simplifies variable management, and reduces coding errors, resulting in faster time to market
Video Player is loading.
Current Time 0:00
Duration 25:11
Loaded: 0.66%
Stream Type LIVE
Remaining Time 25:11
 
1x
  • Chapters
  • descriptions off, selected
  • captions off, selected
  • en (Main), selected
    Video length is 25:11

    Haier engineers are developing leading smart appliances such as washing machines, refrigerators, and air conditioners. However, given the increasing complexity of embedded software in these products, which can support more fault detection, AI, and IoT functions, traditional R&D development models face many difficulties, including high costs, delayed time to market, and the inability to systematically manage the increasingly large number of software and hardware design artifacts.

    To improve R&D efficiency, Haier engineers have chosen Model-Based Design. Take the development of a variable frequency motor drive module of a washing machine as an example. The sensorless control algorithm was modeled using Simulink® libraries, including field-oriented control, position and speed estimation, maximum torque per ampere, space vector pulse width modulation, and other functional modules. Using Stateflow® made it easy to model the appliance control logic, including ADC offset check, initialization, startup, operation, failure, and other states, as well as cumbersome jump management.

    To facilitate the design and optimization of control algorithms, engineers modeled virtual plants—including inverters and motors—using Simscape Electrical™, making it possible to verify the algorithms without hardware. After this verification step, Embedded Coder® was used to generate production-ready code and integrate it with other legacy handwritten code. Automatic code generation avoids possible manual coding errors and improves the efficiency of hardware deployment.